Background/Objective: Breast density and cancer risk are key imaging-derived biomarkers, yet their assessment is limited by inter-reader variability and inconsistent reproducibility. This Technical Note evaluates the feasibility of a bifurcated neural network designed to simultaneously predict breast density and a composite cancer risk index, providing a methodological foundation for future integration into contrast-enhanced mammography (CEM) workflows. Materials and Methods: A simulated cohort of 1000 patients was generated to reproduce clinically plausible variability in breast density (Densitanum) and cancer risk (RiskEnum). A multi-output neural network was developed and compared with two baselines: multiple linear regression and a single-output multilayer perceptron (MLP). Performance was assessed using R2, mean squared error (MSE), and mean absolute error (MAE). Learned trends were examined for consistency with established physiological and epidemiologic patterns. Results: Linear regression showed limited explanatory power (R2 ≈ 0.144). The single-output MLP improved prediction of the cancer risk index (R2 = 0.436; MSE = 9.558). The bifurcated neural network achieved MAE values below 4 for both outputs (2.624 for Densitanum; 3.731 for RiskEnum), demonstrating robust performance and the advantage of simultaneous multi-target prediction. The model reproduced clinically coherent patterns, including the expected age-related decline in breast density. Conclusions: This simulation-based feasibility study demonstrates that bifurcated neural networks can jointly model correlated breast imaging biomarkers with high internal consistency. The proposed architecture provides a reproducible methodological platform that can be directly tested on real CEM datasets to support future AI-enhanced risk stratification and personalized screening pathways.
18 December 2025 in Eurasian Journal of Medicine and Oncology
AccScience Publishing is a publishing company based in Singapore. We have in our portfolio a range of high-quality, open-access, peer-reviewed journals and books from a broad spectrum of science, technology, biomedicine, engineering, education, social sciences and many more.
Background: Breast density (BD) and background parenchymal enhancement (BPE) have long been considered related descriptors of breast tissue composition, yet growing evidence shows that they represent distinct and non-interchangeable physiological domains. Objective: This study aims to evaluate whether BD and BPE represent independent imaging biomarkers with distinct systemic associations, and to investigate their potential role in preventive medicine beyond breast cancer detection. Methods: Drawing on linear modeling, neural analyses, inverse prediction, and correlation studies conducted across five coordinated investigations, we evaluated the systemic associations of BD and BPE. We also performed model ablation analyses to confirm biomarker independence and assessed the impact of AI support on inter-reader agreement and workflow efficiency. Results: BD is strongly associated with bone mineral density, supporting its role as a structural marker linked to osteoporotic risk. Conversely, BPE correlates with systolic blood pressure (SBP), reflecting vascular tone and perfusion. Importantly, the systemic information carried by BD can be extracted not only from contrast enhanced mammography (CEM) but also from standard mammography. When CEM is available, BPE adds complementary insight into vascular physiology. Furthermore, AI support improved inter-reader agreement and workflow efficiency. Conclusions: These findings support a dual biomarker framework in which BD and BPE independently contribute to systemic risk stratification. This approach suggests potential for opportunistic osteoporosis risk assessment from routine screening examinations, though prospective validation is required before clinical implementation. This framework may shift breast imaging toward a broader, sustainable, and human-centered approach to preventive health.